Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A document image processing apparatus for use with a document image comprising a plurality of character string element regions, the document image processing apparatus comprising: a memory for storing the document image; and a controller for controlling extraction of an index region from said document image, wherein said controller is configured to: for each character string element region, identify the character string element region as a large region or a small region based on at least one of a size property and a character property of the character string element region; for each small region, identify the small region as an index candidate when a region immediately following the small region is a large region, and identify the large region immediately following the index candidate region as a related text region; for each index candidate region, compare formatting of the index candidate region to formatting of the related text region and identify a formatting property that is different between the index candidate region and the related text region as a different formatting property; for each index candidate region, set an evaluation criterion such that a value of the index candidate region for the different formatting property satisfies the evaluation criterion and a value of the related text region for the different formatting property does not satisfy the evaluation criterion; for each index candidate region: set the index candidate region as the focused index candidate region and set the different formatting property of the focused index candidate region as the focused different formatting property; calculate at least one of a number of similar index candidate regions and a number of similar large regions, wherein: the number of similar index candidate regions is a total number of index candidate regions among the index candidate regions, except for the focused index candidate region, that satisfy the evaluation criterion for the focused different formatting property; and the number of similar large regions is a total number of large regions that satisfy the evaluation criterion for the focused different formatting property; calculate an index evaluation degree based on at least one of the number of similar index candidate regions and the number of similar large regions; identify index regions from among the index candidate regions based on the index evaluation degree of each index candidate region.
A document image processing system extracts index regions from document images containing text. It identifies text regions as either "large" or "small" based on size and character properties. If a small region is immediately followed by a large region, the small region is considered an index candidate and the large region is its related text. The system then compares the formatting of the index candidate and related text, noting any differences (e.g., font, indentation). It sets an evaluation criterion based on these formatting differences, such that only the index candidate satisfies it. It calculates how many other index candidates and large regions meet this criterion. Finally, it uses these counts to determine an index evaluation score, identifying actual index regions based on this score.
2. The document image processing apparatus according to claim 1 , wherein the index evaluation degree is such that as the number of similar index candidate regions increases, a likelihood of identifying the index candidate region as an index region increases; and as the number of similar large regions increases, the likelihood of identifying the index candidate region as an index region decreases.
The document image processing system described previously calculates an index evaluation score based on the number of similar index candidates and similar large regions. The index evaluation score increases the likelihood of identifying a region as an index region as the number of similar index candidates increases, and decreases the likelihood of identifying a region as an index region as the number of similar large regions increases. This means more similar index candidates make it more likely to be an index, while more similar large regions make it less likely.
3. The document image processing apparatus according to claim 2 , wherein said controller identifies the index candidate region as an index region when the number of similar large regions is a predetermined number or less.
The document image processing system from the previous description, which identifies index regions based on the number of similar index candidates and large regions, specifically identifies an index candidate region as an index region only if the number of similar large regions is below a certain predetermined threshold. This prevents incorrect index identification if there are many large regions with similar formatting to the index candidate.
4. The document image processing apparatus according to claim 2 , wherein said controller identifies the index candidate region as an index region when the number of similar index candidate regions is a predetermined number or more.
The document image processing system, identifying index regions based on similar index candidates and large regions, identifies an index candidate region as an index region if the number of similar index candidate regions is above a certain predetermined threshold. This confirms a pattern and strengthens the likelihood that it is truly an index region.
5. The document image processing apparatus according to claim 1 , wherein when the index evaluation degree cannot be calculated, the evaluation criterion is changed.
In the document image processing system, if the initial calculation of the index evaluation score fails (e.g., due to insufficient data or an ambiguous evaluation criterion), the system changes the evaluation criterion. This ensures the system can adapt and still identify index regions even when the initial assumptions are not met.
6. The document image processing apparatus according to claim 5 , wherein said controller identifies a new formatting property different from the different formatting property, and sets a new evaluation criterion for the new formatting property, as part of the change of the evaluation criterion, and calculates the number of similar index candidate regions and the number of similar large regions based on both the evaluation criterion and the new evaluation criterion.
When the index evaluation score cannot be calculated in the document image processing system, the system changes the evaluation criterion by identifying a new formatting property different from the initially considered formatting property and sets a new evaluation criterion for this property. The system then recalculates the number of similar index candidate regions and similar large regions, using both the original and the new evaluation criteria to refine the index identification process.
7. The document image processing apparatus according to claim 1 , wherein said different formatting property comprises any one of an indent amount, a justified line, distances from previous and next character string element regions, a character string decoration, a head character kind, an end character kind, a language type, a character size, a line height, a character interval, a font, a character decoration, a character color, and a background color.
The "different formatting property" used by the document image processing system for index candidate evaluation can include various characteristics like indent amount, justified line, distance from surrounding text, text decorations (e.g., bold, italics), beginning/ending characters, language, character size, line height, character spacing, font type, character decoration, character color, or background color. The system uses any of these formatting differences to distinguish index candidates from related text.
8. The document image processing apparatus according to claim 1 , further comprising an image reader to read a manuscript to generate said document image, wherein said controller further generates viewing navigation information for specifying a position of said extracted index region in said document image, and generates a computerized document by adding said viewing navigation information to said document image.
The document image processing system also includes an image reader to scan physical documents and create the digital document image. After extracting index regions, the system generates navigation information that links the extracted index regions to their corresponding locations within the document image. The system then combines the processed image with the viewing navigation information to create an enhanced, searchable document.
9. The document image processing apparatus of claim 1 , wherein the apparatus is a multi-function peripheral.
The document image processing system is implemented as a feature of a multi-function peripheral (MFP) device, such as a printer or scanner.
10. The document image processing apparatus of claim 1 , wherein the size property is character size or line height.
In the document image processing system, the "size property" used to distinguish between large and small text regions is either the character size or the line height of the text.
11. The document image processing apparatus of claim 1 , wherein the evaluation criterion is dynamically set with respect to each index candidate region.
The evaluation criterion used by the document image processing system is dynamically set with respect to each index candidate region. This means the system adjusts the specific criteria based on the unique context of each potential index, improving accuracy.
12. The document image processing apparatus according to claim 1 , wherein the different format property indicates that index candidate region includes one of format properties and the related text region does not include the format property, and the evaluation criterion is the criterion for evaluating whether the format property is existing in the region or not.
In the document image processing system, the "different format property" indicates that an index candidate region has a certain formatting property (e.g., bold font) that the related text region does not have. In this case, the evaluation criterion is based on whether the formatting property is present in the region or not. The system simply checks for the existence of this formatting property.
13. A document image processing method that extracts an index region from a document image comprising a plurality of character string element regions, the method comprising: providing a document image processing apparatus comprising a memory for storing the document image and a controller for controlling extraction of an index region from the document wherein said controlling extraction of an index region from the document image comprises: for each character string element region, identifying the character string element region as a large region or a small region based on a size property of the character string element region; for each small region, identifying the small region as an index candidate when a region immediately following the small region is a large region, and identifying the large region immediately following the index candidate region as a related text region; for each index candidate region, comparing formatting of the index candidate region to formatting of the related text region and identifying a formatting property that is different between the index candidate region and the related text region as a different formatting property; for each index candidate region, setting an evaluation criterion such that a value of the index candidate region for the different formatting property satisfies the evaluation criterion and a value of the related text region for the different formatting property does not satisfy the evaluation criterion; for each index candidate region: setting the index candidate region as the focused index candidate region and setting the different formatting property of the focused index candidate region as the focused different formatting property; for each index candidate region that is not the focused index candidate region, identifying the value of the index candidate region for the focused different formatting property; calculating, as a number of similar index candidate regions, a total number of index candidate regions that satisfy the evaluation criterion for the focused different formatting property; for each large region, identifying the value of the large region for the focused different formatting property; calculating, as a number of similar large regions, a total number of large regions that satisfy the evaluation criterion for the focused different formatting property; calculating an index evaluation degree based on the number of similar index candidate regions and the number of similar large regions; identifying index regions from among the index candidate regions based on the index evaluation degree of each index candidate region.
A document image processing method extracts index regions from a document image containing multiple character string element regions. It identifies text regions as either "large" or "small" based on size. If a small region is followed by a large region, the small region is an index candidate and the large region is its related text. The formatting of the index candidate is compared to the related text to find differences. An evaluation criterion is set based on these formatting differences so that it is satisfied by the index candidate, but not by its related text. For each index candidate, the algorithm calculates the number of similar index candidates and similar large regions that satisfy the criterion. Finally, it identifies index regions based on an index evaluation score calculated from these counts.
14. The document image processing apparatus according to claim 13 , wherein the different format property indicates that index candidate region includes one of format properties and the related text region does not include the format property, and the evaluation criterion is the criterion for evaluating whether the format property is existing in the region or not.
In the document image processing method, the "different format property" indicates that an index candidate region has a certain formatting property (e.g., bold font) that the related text region does not have. The evaluation criterion is based on whether the formatting property is present in the region or not. The method simply checks for the existence of this formatting property.
15. The document image processing apparatus according to claim 13 , wherein said different formatting property comprises any one of an indent amount, a justified line, distances from previous and next character string element regions, a character string decoration, a head character kind, an end character kind, a language type, a character size, a line height, a character interval, a font, a character decoration, a character color, and a background color.
In the document image processing method, the "different formatting property" used for index candidate evaluation can include various characteristics like indent amount, justified line, distance from surrounding text, text decorations (e.g., bold, italics), beginning/ending characters, language, character size, line height, character spacing, font type, character decoration, character color, or background color. The method uses any of these formatting differences to distinguish index candidates from related text.
16. A non-transitory computer-readable recording medium having a recorded document image processing program for use with a document image comprising a plurality of character string element regions that, when executed by a computer, causes the computer to perform: for each character string element region, identifying the character string element region as a large region or a small region based on a size property of the character string element region; for each small region, identifying the small region as an index candidate when a region immediately following the small region is a large region, and identifying the large region immediately following the index candidate region as a related text region; for each index candidate region, comparing formatting of the index candidate region to formatting of the related text region and identifying a formatting property that is different between the index candidate region and the related text region as a different formatting property; for each index candidate region, setting an evaluation criterion such that a value of the index candidate region for the different formatting property satisfies the evaluation criterion and a value of the related text region for the different formatting property does not satisfy the evaluation criterion; for each index candidate region: setting the index candidate region as the focused index candidate region and setting the different formatting property of the focused index candidate region as the focused different formatting property; for each index candidate region that is not the focused index candidate region, identifying the value of the index candidate region for the focused different formatting property; calculating, as a number of similar index candidate regions, a total number of index candidate regions that satisfy the evaluation criterion for the focused different formatting property; for each large region, identifying the value of the large region for the focused different formatting property; calculating, as a number of similar large regions, a total number of large regions that satisfy the evaluation criterion for the focused different formatting property; calculating an index evaluation degree based on the number of similar index candidate regions and the number of similar large regions; identifying index regions from among the index candidate regions based on the index evaluation degree of each index candidate region.
A computer program stored on a non-transitory medium extracts index regions from a document image. It identifies text regions as either "large" or "small" based on size. If a small region is followed by a large region, the small region is an index candidate and the large region is its related text. The formatting of the index candidate is compared to the related text to find differences. An evaluation criterion is set based on these formatting differences so that it is satisfied by the index candidate, but not by its related text. For each index candidate, the algorithm calculates the number of similar index candidates and similar large regions that satisfy the criterion. Finally, it identifies index regions based on an index evaluation score calculated from these counts.
17. The document image processing apparatus according to claim 16 , wherein the different format property indicates that index candidate region includes one of format properties and the related text region does not include the format property, and the evaluation criterion is the criterion for evaluating whether the format property is existing in the region or not.
In the document image processing program, the "different format property" indicates that an index candidate region has a certain formatting property (e.g., bold font) that the related text region does not have. The evaluation criterion is based on whether the formatting property is present in the region or not. The program simply checks for the existence of this formatting property.
18. The document image processing apparatus according to claim 16 , wherein said different formatting property comprises any one of an indent amount, a justified line, distances from previous and next character string element regions, a character string decoration, a head character kind, an end character kind, a language type, a character size, a line height, a character interval, a font, a character decoration, a character color, and a background color.
In the document image processing program, the "different formatting property" used for index candidate evaluation can include various characteristics like indent amount, justified line, distance from surrounding text, text decorations (e.g., bold, italics), beginning/ending characters, language, character size, line height, character spacing, font type, character decoration, character color, or background color. The program uses any of these formatting differences to distinguish index candidates from related text.
Unknown
September 16, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.